Hi there--
I want to run a paired t-test to analyse the effect of a treatment, but I have some problems to define the contrats.
To analyse the effect of the treatment I performed a F contrast: (1 -1), is it right?
and how can i define the effect of interest?
basically I have two levels (pre post) and 20 subjects. this is the spm's matrix:
Subj 1-> 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Subj 1-> 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Subj 2-> 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Subj 2-> 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
......
Subj 20->0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
Any advice would be greatly appreciated, Also eventually reading suggestions.
Thanks!
Andrea
|